Billionaires and Stealth Politics Read online

  that they likely care a great deal about, such as taxes and Social Security.

  But we also expected that— in line with the well- documented tendency

  of affluent and wealthy citizens to participate in politics at particularly high levels27— many billionaires would take political action directly related to these policy issues: that, for example, they would give money to policy- oriented causes and would hold fund- raisers and bundle contributions by others.

  We did expect to find systematic variations among billionaires, how-

  ever. The level of wealth that people enjoy tends to affect their policy

  preferences. For example, more- affluent people tend to be more conser-

  vative on various economic issues. One possibility is that the monotonic

  association of greater affluence with greater economic conservatism con-

  tinues right to the top, even among billionaires who hold smaller or larger fortunes. A different possibility is that as billionaires’ wealth rises, their policy preferences may at some point actually begin to move in the opposite direction. Some very high level of wealth may be felt to be “enough.”

  Or, more precisely, if the subjective marginal value of wealth declines as wealth increases, while considerations related to altruism, social pressure, or the provision of public goods remain fixed or grow, then at some point the perceived net marginal costs of high taxes and high social welfare

  spending may begin to tip over into perceived net benefits.

  Given the likelihood that policy preferences vary with wealth even

  within this superwealthy group, we expected to find political silence more common at wealth levels where billionaires’ views are most at odds with

  the views of the general public. We also expected that— controlling for

  wealth level— the inheritors of great wealth and billionaires who are most directly exposed to consumers would be least likely to speak out on the

  issues.

  Studying the Political Words and Actions of Billionaires

  Our study concerns the very wealthiest one hundred Americans, the top

  one hundred denizens of the Forbes 400 list as of October 2013: from Bill Gates at the top (with $72 billion in net worth), to Daniel and Dirk Ziff at the bottom (with just $4.6 billion each).28 All one hundred billionaires taken together had a combined net worth of about $1,291 billion (that is, about one and one- quarter trillion dollars— more than the entire annual

  stealth politics on taxes and social security

  33

  GDP of Mexico or the Netherlands, and only slightly less than that of Aus-tralia or South Korea).29 All one hundred are listed in appendix 3, along with their net worth and their scores on some of our dependent variables.

  It is not possible to ascertain billionaires’ policy preferences by survey-ing a representative sample of them.30 Even the Survey of Consumer Fi-

  nances by the National Opinion Research Center (NORC)— the only ex-

  isting survey that regularly interviews a representative sample of wealthy Americans, which is conducted for the Federal Reserve Board and is

  known for the high level of cooperation it elicits— does not attempt to

  contact Forbes 400 billionaires.31 They are simply too busy and too protective of their privacy to survey. Nor do we believe that one can reliably infer billionaires’ policy preferences from their contributions to political candidates and parties,32 since donations made to moderate or opposite-party officials for nonideological reasons like cultivating access may make the donors look more “moderate” than they actually are.

  It is relatively straightforward, however, to ascertain the publicly stated issue stands of billionaires or of any other prominent individuals. In these days of ubiquitous electronic media, careful online searches can reveal

  virtually every utterance that they make in public. Formal speeches or

  videos, op- ed columns, letters to the editor, journalistic interviews— even offhand or overheard comments— can generally be dredged up.

  For this chapter, we focus on what the one hundred wealthiest billion-

  aires have said, over a roughly ten- year period, about specific policies related to taxes and Social Security. On tax policy, we scrutinized everything they said about capital gains rates; corporate tax rates; the estate tax; the earned income tax credit; carbon taxes; the charitable tax deduction; the so- called Buffett Rule; flat tax proposals; a financial transaction tax; and, more generally, tax revenue expansion or wealth redistribution. On Social Security, we examined everything the billionaires said about payroll tax rates; the payroll tax cap; Social Security privatization; means testing; the retirement age; and, more generally, Social Security reform and benefit reductions.33

  Web Scraping and Search Terms

  Separately for each of the one hundred wealthiest billionaires, we carried out a systematic internet- based search for all statements that could be

  found concerning specific policies related to taxation or Social Security policy.

  34

  chapter two

  A crucial aspect of this research involved the careful development of

  a refined set of policy- related search terms: terms that would capture all statements that are relevant to our policy issues but not produce such an overwhelming number of irrelevant hits that the wheat gets lost in chaff.

  In theoretical terms, the objective is to map the “semantic field”34 associated with each policy domain: that is, the family of terms actually used by billionaires and others to refer to that domain as a whole or to specific parts of it. The semantic field includes language employed by social scien tists, but also the sometimes quite different terminology employed by diverse

  supporters or opponents of various alternatives in a given policy domain.

  It is helpful to imagine the semantic field as a network, in which a con-

  cept in a domain is linked with another term in that domain if they are

  often used together. For example, the terms “estate tax” and “death tax”

  should be linked, because they are both likely to occur in a number of

  written or spoken texts concerning taxes that are applied to the prop-

  erty of wealthy individuals at the time of their death. However, terms

  that relate to the same policy domain but have quite different ideological connotations may not be directly connected; they may instead have one or

  more steps of separation. For example, not many texts are likely to refer to Social Security using both the language of the “social safety net” and references to “generational theft.” Thus, these terms would lack a direct link in the conceptual network. Nonetheless, they would be indirectly

  linked at one step of separation because both co- occur regularly (but not always) with the phrase “Social Security.”

  Our search procedure started with the most obvious and visible term

  for each policy domain (e.g., “tax[es]” or “tax[ation]” and “Social Secu-

  rity”).35 Using that term as a keyword, we carried out Google News/gen-

  eral web and LexisNexis searches not restricted to the words of billion-

  aires or any other particular individuals. We read the top several hits on each search and noted the alternative terms that were used in US public

  discourse (as exemplified by those texts) to refer to the relevant policy domain. We then repeated the process iteratively, using each prominent

  search term that was uncovered in the previous search. Because we fol-

  lowed each term connected with the initial term for at least five search

  steps, rather than predetermining a target number of terms to character-

  ize each domain, the process is more similar to a series of random walks

  than to a traditional snowball sample. As a result, it is reasonable to expect that our sample of search keywords replicates the topological propertie
s of the original network— and, of greater substantive interest for this

  stealth politics on taxes and social security

  35

  proj ect, captures the most- connected nodes.36 Hence there is good reason to believe that our process has captured a considerable proportion of the actually existing political terminology connected with each policy domain.

  Some results of this iterative procedure were predictable. For example,

  in addition to searching for statements made about the “estate tax,” we

  were also led to search for statements about the “death tax.” In addition to searches related to support or opposition to “tax increases,” we also

  searched for positions on “revenue enhancement.” We ultimately settled

  on twenty- four keywords related to taxation and ten keywords related to

  Social Security (see appendix 2).

  After developing this comprehensive list of keywords, we turned to the

  web to begin searching.37 We originally used two main resources: Google

  News/general web search and LexisNexis Academic search. Google and

  LexisNexis provide complementary resources. Google produces a very

  large number of potentially relevant web pages and, helpfully, includes

  links to videos of interviews with our subjects. Google’s search results, however, are sometimes noisy and include numerous websites of dubious

  authority or pronounced ideological predispositions.38 LexisNexis pro-

  duces a smaller number of results and does not include video links, but

  it draws exclusively from mainstream journalistic and academic sources.

  After collecting data for approximately one- quarter of the sample, how-

  ever, we discovered that LexisNexis searches did not uncover any relevant political statements missed by Google searches. After this discovery, we

  decided to use Google exclusively.

  Once we began conducting web searches, we quickly realized that some

  of the Forbes billionaires who are widely believed to be very active politically are nonetheless rather tight- lipped when it comes to discussing politics. The Koch brothers (David and Charles) are a leading, although

  an imperfect (and now, ironically, a celebrated), example of this. While they have occasionally spoken in public about their political views, and

  while— in recent years, with help from investigative journalists— most

  politically informed Americans have become aware of the general thrust

  of their political stands, the brothers nonetheless speak publicly a great deal less often than they make large financial contributions to political causes. And when they do speak out, they tend not to take specific policy stands. Sometimes they even deliberately misrepresent the main thrust of

  their preferred policies, as when Charles Koch insisted that he “agree[d]”

  with Bernie Sanders on several matters.39 This combination is suggestive

  of what we call “stealth politics.”

  36

  chapter two

  Of course the most famous examples of stealth politics, like the

  Kochs, are far from the best examples. As we will see, many other bil-

  lionaires have been just as quiet but much more successful at stealth

  politics. They remain obscure even to citizens who pay a lot of attention to politics.

  Political Actions

  In order to study actions as well as words, and to further explore our hypotheses about stealth politics, we also included in our search highly spe cific, issue­ oriented actions, including financial contributions to issue­ specific organizations. The Center for Responsive Politics, which runs the OpenSecrets.org website, is a very helpful resource on these matters. Our re-

  search included a search for each of the Forbes billionaires in OpenSecrets’

  online database of reported contributions to candidates and political ac-

  tion committees (PACs). Though some existing PACs are candidate-

  rather than policy- specific, many are narrowly and explicitly focused on a small set of specific issues. Donations to these sorts of PACs are included in our raw data on political actions because their specific policy content is easy to measure. We also took note of board and advisory positions at

  policy- specific organizations.40

  We did not include contributions to candidates, parties, or candidate-specific PACs as policy- specific actions. For the reasons noted above, and because parties and candidates typically take stands across a range of unrelated issues, we are not convinced that they can be reliably used to measure the goals of billionaires’ political activity. We did, however, gather Fed eral Election Commission (FEC) data on the billionaires’ total party

  and candidate contributions, which are reported below. We also separately searched for media reports of instances in which the billionaires served as bundlers of political contributions or hosted political fund- raisers.

  Our pursuit of policy- oriented actions as well as words was complicated

  by the exemption of various types of 501(c) organizations from the man-

  datory reporting of financial contributions to which explicitly political organizations are ordinarily subjected. As a result, our search had to rely to some extent on investigative efforts by journalists and open- government

  organizations to uncover dark- money contributions. Again, the Center

  for Responsive Politics, particularly in its collaboration with the Washington Post, proved to be a helpful resource. Our search inevitably missed many dark- money contributions that were funneled through certain types

  stealth politics on taxes and social security

  37

  of 501(c) organizations— no surprise, since one point of dark- money con-

  tributions can be to hide them— but by this procedure we were nonethe-

  less able to identify some additional policy- specific political actions. To the extent that certain particularly secretive contributions were missed, any findings of stealth politics are likely to be under stated, not overstated.

  Although our web scraping followed a carefully defined process and was

  as methodical as possible, it inevitably required some exercises of judg ment.

  Sometimes journalists’ accounts of speeches and interviews alluded to policy positions without including direct quotations. In such cases, we had to search for transcripts or videos, or settle for paraphrases of stands that were confirmed by other journalists’ reports. Other times, search results included links to discussions of interviews and speeches on blogs or news aggregation sites. These sites typically drew from accounts by traditional journalists, which we had to track down before we could search for transcripts or videos.

  This search process was often arduous.41

  Since we are interested in silence as well as voices, we were careful to

  use numerous different keywords related to each policy topic to search

  several outlets, even when we had reason to believe that a particular billionaire had not taken any public political positions. The payoff for scrupulously careful searching is that, when we failed to find political statements, we could be reasonably confident that no such statements were made.

  Coding

  After compiling all tax- and Social Security– related words and actions for each of our one hundred billionaires, we produced measures on three sets

  of dependent variables: (1) a simple count of the number of public statements made by each billionaire on each policy subissue and on tax and Social Security policy as a whole (together with a dichotomous variable indicating presence or absence of at least one statement on the issue or subissue); (2) a dichotomous variable indicating presence or absence of political action on each issue or subissue (both overall, and separating personal issue-oriented financial contributions or memberships from bundling or raising

  contributions from
others); and (3) measures of the liberal- conservative directionality of each billionaire’s statements and of his or her political actions on each subissue (and on all tax and all Social Security issues taken together).

  The measures for each billionaire were aggregated in various ways to

  produce frequency distributions and to test our expectations about varia-

  tion by wealth level and source of wealth.

  38

  chapter two

  The Silence of the Lions

  One of the most important findings from our web scraping concerns the

  extent to which the wealthiest American billionaires have spoken out

  in public concerning taxes or Social Security. They have very seldom

  done so.

  On tax policy, only twenty- six of the one hundred wealthiest billion-

  aires— that is, only about one- quarter of them— made any public state-

  ment at all concerning whether any specific tax— or taxes in general—

  should be raised, lowered, abolished, or kept the same. Despite the fact

  that tax policy is highly important and had been vigorously debated in

  public on numerous occasions, nearly three- quarters of the billionaires

  were entirely silent, or made only vague statements (for example, that a

  policy domain is “important”) that completely lacked a liberal, conserva-

  tive, or centrist policy direction.

  This is not for lack of opportunity. The billionaires had plenty of time

  to speak; our searches reached back at least ten years.42 They had plenty of opportunity to speak. (Again, many journalists would be delighted to interview billionaires.) But most chose silence. (See table 2.1 and appendix 3.) Our figure of about one- quarter (26 percent) of the billionaires speaking out on taxes may actually tend to give an inflated impression of the

  extent to which they took stands, because it is aggregated over all the different types or subissues of tax policy that we examined. Taking specific subissues one by one, only 19 percent of the billionaires said anything at all about personal income tax rates; only 13 percent about corporate tax

  rates; just 8 percent about the estate tax; a mere 6 percent about carbon taxes; and a similarly meager 5 percent about capital gains taxes.